xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-01-23 11:24:31 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Performance][gpt-oss] Revert gpt-oss max cudagraph size to 1024 (#28345)

Browse Source 
Signed-off-by: Mohammad Miadh Angkad <MAngkad.BSDSBA2027@aim.edu>
This commit is contained in:
Mohammad Miadh Angkad 2025-11-09 06:50:10 +08:00 committed by GitHub
parent 171133f929
commit 404d7a9d14
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 4 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
vllm/model_executor/models/config.py
View File

@ -258,9 +258,9 @@ class GptOssForCausalLMConfig(VerifyAndUpdateConfig):
if structured_outputs_config.reasoning_parser == "":
structured_outputs_config.reasoning_parser = "openai_gptoss"
# Increase the max capture size from 512 to 992 for performance.
# Increase the max capture size from 512 to 1024 for performance.
# NOTE(woosuk): This will increase the number of CUDA graphs
# from 67 to 81.
# from 67 to 83.
compilation_config = vllm_config.compilation_config
# Only override when the user has not set either of
# cudagraph_capture_sizes or max_cudagraph_capture_size.
@ -268,11 +268,9 @@ class GptOssForCausalLMConfig(VerifyAndUpdateConfig):
compilation_config.cudagraph_capture_sizes is None
and compilation_config.max_cudagraph_capture_size is None
):
# FIXME(woosuk): When using full cuda graph with FA3, the max
# supported size is 992.
compilation_config.max_cudagraph_capture_size = 992
compilation_config.max_cudagraph_capture_size = 1024
logger.info(
"Overriding max cuda graph capture size to %d for performance.", 992
"Overriding max cuda graph capture size to %d for performance.", 1024
)